using new technology on old machines. Was: PDP-12 Restoration at the RICM

[Jay Jaeger]

Not all EE's have the same education with regard to how semiconductors
function.  When I was in school I took a class in semiconductor physics
- an entire semester on how the wee beasties function - more than most EEs.

The prof., Henry Guckel, told an interesting story about an advanced IBM
computer he did some work on (not sure if it was the ACS or one of the
9x models).  The clock ran really fast, and they were had all kinds of
problems with propagation delays - for which their typical cure was to
add delay lines - to the point where it slowed down the machine so much
that they actually had to back off the clock some. He also didn't like
computer programmers very much.

Another related story: a pre-requisite for the class was a modern
physics class - which was horribly taught, by a prof. the physics
department.  During the 2nd class or so, Prof. Guckel noticed a lot of
"deer in headlights" looks, and stopped.  After commenting that the
material should have been largely review, he asked when and from whom we
had taken the modern physics class - and 3/4 of the class had taken the
class from the same prof. during the previous 12 months.  After
realizing that class had been a disaster, Prof. Guckel spent the next 4
lectures re-teaching most of that entire-semester class - and doing a
really good job of it - preparing his own notes and everything, before
proceeding through the rest of the class.  Really tough course.

On 7/13/2015 6:18 AM, Tor Arntsen wrote:

> 
> I think I'm reasonably well into the pragmatic camp (ref. Chuck G.'s
> post). But one point about 555s and Arduinos is that I couldn't build
> a 555 more easily than I could build the MCU on an Arduino board (a
> more relevant comparision might be a 555 vs a Propeller chip) -
> they're all black boxes to me. Even a transistor, to some extent - I
> know exactly how it works, well, as much as any other EE anyway, but I
> couldn't build one. Or a vacuum tube..
>